I can advise several methods: Correlation Linear r (Pearson) parametric correlation coefficient. Spearman's D and rs, Spearman's (non-parametric) rank-order correlation coefficient is the linear correlation coefficient (Pearson's r) of the ranks. Kendall's tau non-parametric correlation coefficient. For these analyzes a Permutation tests Monte Carlo permutation tests (N = 9999) are available for all the correlation coefficients in the Past.3.18. You can also perform a test table (Mantel 1967, Mantel & Valand 1970) is a permutation test for correlation between two distance or similarity matrices. You can also perform linear bivariate models: Ordinary Least Squares (OLS), Reduced Major Axis (RMA), Major Axis (MA), and Robust or use as suggested by Dr. Kumar some regression method.

Attached the manual of Past. 3.18 to analyze which method is best suited to your variables and work objectives.

Instead of using the means of water absorbed and the means of time please plot the raw data in a scatter plot and see how it looks. If there is a linear relationship as evident from the graph fit a simple linear regression model. Excel can do this easily. Go to Chart-> Insert->Scatter Plot and after it displays the data as symbols ask it to fit a trend line (linear first). Write down the R-square value. If you see curvature then fit a quadratic trend and again note the R-square value.

Dear Srini Vasan and Ajit kumar Roy

Thank you very much for these wonderful suggestions. I have tried the scatter plot and it is giving two parallel sets od dots and when I ask the trend line it gives trend lines based on a set of variables: the line for absorbed water is below since values are very small and the line for cooking time is up in a different color.

Could you guide more, Sirs.

Thank you very much.

Thanks!

Dear Enrique Reynaldo de la Cruz

Thank you very much for the wonderful suggestion. I am still analysing the various corrélation methods you have proposed and try to understand. The document you attached is very rich, I am exploiting it. Thank you very much.

Please, is PAST free for download? I am using XLTAT but it looks like that PAST described in the manual is more friendly.

Thanks

Regards!

Dear Srini Vasan and Ajit kumar Roy

I have tried the scatter plot on a different set of data and obtained the attached diagram; the trend seems to show ''NO CORRELATION'', please, can you advise on the next type of analysis to run? it give a correlation coefficient of -0.0395.

Any more advise sirs!

Thank you very much for your understanding and sharing.

Looking forwards to hearing from you soon.

Regards!

The value of R is close to zero, it means that there is no relationship between both variables, in addition there is a high dispersion of the results. I recommend you perform a logarithmic transformation of both variables and run the test again. I recommend you also use the Mantel test for 10,000 interactions if you repeat the result is that there is no relationship between both variables.

Dear Difo Voukang Harouna Difo Voukang Harouna

Sorry for the delay in viewing the graph and replying. Yes there is no correlation. I agree with Dr. de la Cruz above. Transform the independent variable using logarithm and see if the R value goes up and the scatter comes down.There are other types of transformations also available ( https://en.wikipedia.org/wiki/Data_transformation_(statistics) ) including sine functions. I think you may have to repeat the data collection and see what causes the scatter and try to minimize it. Look for the extreme outliers and see what causes them. Best wishes

Correlation matrix .check up on Stata Software